Subintimal angioplasty for long (TASC C and D) superficial femoral artery occlusions

(1)

ORIGINAL ARTICLE

Subintimal angioplasty for long (TASC C and D)

superﬁcial femoral artery occlusions

Hassan Abdelsalam

a,*

_{, Tze Yuan Chan}

a

_{, Alistair Millen}

b

_{, Robert Fisher}

b

a

Radiology Department, Royal Liverpool University Hospital, UK

b_{Vascular Surgery Department, Royal Liverpool University Hospital, UK}

Received 12 January 2015; accepted 26 May 2015 Available online 23 June 2015

KEYWORDS Subintimal; Angioplasty; SFA;

Long occlusion

Abstract Sub-intimal angioplasty (SIA) is a recognized endovascular option for long segment occlusive disease. The Inter-Society Consensus for the Management of peripheral arterial disease (TASC) published a revised classiﬁcation for infra-inguinal peripheral vascular disease in 2007. We present our experience with this technique and aim to determine the efﬁcacy of sub-intimal angioplasty (SIA) for TASC C and TASC D category lesions.

Materials and Methods: A retrospective analysis of all consecutive SIA for TASC C and D super-ﬁcial femoral artery (SFA) occlusions was conducted. The procedures were carried out in a single center between 01/11/2009 and 01/06/2011. Primary endpoints were limb salvage rates. Secondary endpoints were primary, primary assisted, secondary patency, and complication rates. Kaplan– Meier analysis was used to assess the patency, limb salvage and survival rates.

Results: 29 limbs in 26 patients were treated, median age 69 years. Indications included moderate to severe claudication in 18 limbs and critical limb ischemia (CLI) in 11 limbs. There were six TASC C and 23 TASC D lesions with a median lesion length of lesion 27.3 cm (15.2–46.4). Complications occurred in 3 cases (10%), which were all distal embolization. All 3 had successful thrombo-aspirations and were treated with IV heparin for 24–48 h. The technical and clinical success rates were 93.1%. The primary patency rate, primary assisted patency and secondary patency at 1 year of approximately were 41%, 57% and 60% respectively. There was one below knee amputation in the study group, giving a cumulative proportional limb salvage rate of 96.6%. The primary patency of the CLI group was 60% at 1 year.

Conclusion: Although we observed a low primary patency rate there were encouraging limb salvage rates in patients with TASC C and D lesions treated with SIA. It provides an alternative to surgery in high risk patients, with acceptable technical success and complication rates but re-intervention is likely to be required.

Ó 2015 The Authors. The Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).

* Corresponding author at: Department of Radiology, Faculty of Medicine, Alexandria University, Alexandria, Egypt. Tel.: +20 1024647818. E-mail address:hassan_abd_elsalam@yahoo.com(H. Abdelsalam).

Peer review under responsibility of Egyptian Society of Radiology and Nuclear Medicine.

Egyptian Society of Radiology and Nuclear Medicine

The Egyptian Journal of Radiology and Nuclear Medicine

www.elsevier.com/locate/ejrnm

www.sciencedirect.com

http://dx.doi.org/10.1016/j.ejrnm.2015.05.017

0378-603XÓ 2015 The Authors. The Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

(2)

1. Introduction

In 1989, Bolia described the use of subintimal angioplasty (SIA) as an alternative method of lower limb revascularization. Since its description, this technique has become increasingly popular and has a role in the management of peripheral vascu-lar disease. Conventionally, complete total occlusions (CTO) of >10 cm were treated with a surgical bypass. Over the past 2 decades, we have witnessed a change in practice which may be attributable to many factors including the expansion of interventional radiology services and improvement of catheters and balloons. As the scope of SIA as a treatment option for peripheral arterial disease (PAD) widens, the patient selection has also dramatically changed. SIA has been originally described in managing superﬁcial femoral artery (SFA) and popliteal artery (PA) occlusions, and is now applied in iliac and crural arteries. There are numerous published outcome studies since the description of the technique but there is scarce data on long CTO i.e. Trans-atlantic Inter-Society Consensus (TASC II) C and TASC D lesions.

The purpose of this study is to perform a subset analysis of CTOs more than 15 cm treated with SIA in our center.

2. Material and methods

This is a single center retrospective subset analysis of subintimal angioplasty (SIA) performed in our unit. In our institution, all elective vascular cases are discussed in the multi-disciplinary meeting (MDT). Patients are discussed before and after treatment and information is recorded in the vascular database.

All patients who underwent SIA from November 2009 to June 2011 were prospectively recorded in a SIA database. Further procedural information was obtained from the depart-mental radiology information system. All the patients who had SIA were identiﬁed and recruited into this study according to the inclusion and exclusion criteria (Table 1).

Before SIA, all patients had imaging to ensure there was a sufficient inflow into the infrainguinal circulation. The first line investigation was Doppler ultrasound but some patients had CT angiography or MR angiography. All patients were started on an antiplatelet agent and statin when reviewed in clinic or on admission. Full clinical history, physical examination and ankle brachial pressure index (ABPI) were recorded (see

Table 2).

Patients with estimated glomerular ﬁltration rate (eGFR) of less than 30 received pre- and post-procedural intravenous hydra-tion. Diabetic patients on Metformin were converted to another

antiglycaemic agent 24 h before and 48 h after the procedure. Following the procedure, patients with intermittent claudication who were scheduled as a day-case were monitored for 6 h and dis-charged if there were no immediate complications. The majority of patients who had critical limb ischemia (CLI) were inpatients and thus, monitored on a vascular surgical ward.

Following SIA, patients were started on dual anti-platelets, with aspirin and clopidogrel 75 mg once daily each for 6 weeks, if there were no contraindications.

2.1. Procedure

SIA was performed as described by Amman Bolia. The ipsilat-eral antegrade approach is preferable as it provides better wire control and pushability. Contralateral retrograde approach is reserved for cases where an antegrade approach was not pos-sible. The majority of procedures are performed via a 5Fr sheath. 4000 IU of heparin is administered during the proce-dure followed by 1000 IU every subsequent hour.

The subintimal channel is achieved by creating a ‘Bolia loop’ either with a 0.035’’ Radiofocus or with a half stiff J tip hydro-philic wire (Terumo, Tokyo, Japan). The wire was initially sup-ported by a 4 Fr short angled Bolia catheter (Terumo, Tokyo, Japan) followed by a 4 or 5 mm balloon catheter. Once re-entry has been achieved and conﬁrmed, the subintimal channel is immediately dilated with an appropriate balloon. Stenting is only deployed in cases where satisfactory hemodynamics is not achieved on angiography. This is usually due to elastic recoil or signiﬁcant stenosis of >30%. When stenting is required, the entire subintimal channel including entry and re-entry points is stented with a self-expanding stentFig. 1.

2.2. Follow-up

Following SIA, all patients were routinely scheduled for a Doppler ultrasound and reviewed in an outpatient clinic at

Table 1 Inclusion and exclusion criteria.

Inclusion criteria

Clinical status: fontaine > 2a (not inclusive) DSA documentation of CTO of SFA ± PA Subintimal transition of occluded lesion > 15 cm

Discussed in vascular multi-disciplinary meeting and deems SIA as the best management for patient

Exclusion criteria Fontaine 1–2a

History of contrast hypersensitivity Acute limb ischemia

Table 2

Weighting of runoﬀ arteries Site of

intervention

No. of units assigneda

3 2 1

CIA EIA ILA

EIA CFA or SFA PROF

CFA SFA PROF

SFA, POP AT, PT, PER

AT Distal tibial Pedal Arch PT Distal tibial Pedal Arch PER Pedal Arch Collaterals to

AT and PT DP Pedal Arch

CIA = common iliac artery, EIA = external iliac artery, IIA = internal iliac artery, CFA = common femoral artery, SFA = superﬁcial femoral artery, POP = popliteal artery, PROF = profunda femoris artery, AT = anterior tibialis artery, PT = posterior tibialis artery, PER = peroneal artery, DP = dorsalis pedis artery.

a _Points _assigned _for _degree _of _occlusion: _{3 = occluded}

throughout most of its length; 2 = 50–99% stenosed; 1 = 20–49% stenosed; 0 = normal or less than 20 < stenosed.

(3)

Fig. 1 (a and b) Digital subtraction angiography (DSA) showing a short stump of the right SFA, the rest of the SFA and the popliteal artery are occluded, with patent anterior tibial artery as a single run-off vessel. (c–e) Post SIA, recanalization of the SFA and popliteal artery. This patient had patent arteries on a Doppler examination 3 years later.

(4)

6 weeks, 6 months and a year. They have an option to make an appointment in between scheduled appointments whenever symptoms recur. In clinic, pre- and post-procedural ABPI and clinical symptoms were recorded.

If patients failed to attend clinic, we telephoned them to obtain clinical information.

Patients do not receive further scheduled follow-up if they remain asymptomatic after a year. The radiology information system is linked with the death registry enabling us to ﬁnd out whether the patient is deceased even if they were no longer followed-up.

2.3. Deﬁnitions and study endpoints

Primary endpoints for this study were limb salvage rate and primary patency.

Secondary endpoints were technical success, clinical success, primary assisted patency, secondary patency and com-plication rates.

Technical success was deﬁned as successful transition of the target lesion through the sub-intimal space and re-entry into the true lumen, with satisfactory angiographic appearance with no signiﬁcant elastic recoil or residual stenosis of >30%.

Clinical success was deﬁned by the improvement of present-ing symptoms, as judged subjectively by the patient in the case of intermittent claudication or ischemic rest pain and by the improvement or full healing of the ulcer in patients with tissue loss. This was assessed 6 weeks post-procedure and patients were then followed up for a year to assess if they remained symptom free.

Lesion length was measured preferably from CT or MR angiography prior to SIA. In cases where patients only had a Doppler ultrasound as the initial investigation, we used cali-brated angiographic images from the procedure.

Limb salvage is deﬁned as survival without any event of ipsilateral above or below knee amputation.

Primary patency is deﬁned as uninterrupted patency with-out any additional procedure, either surgical or endovascular, of the previously treated lesion.

Assisted primary patency is deﬁned as patency from the time of procedure to the time of occlusion of the previously treated lesion, regardless of the number of procedures in the interim.

SVS scores were recorded according to angiographic appearance as described by Peterkin et al.(1).

Complications are graded to minor and major complica-tions in line with the Society of Interventional Radiology categories (Table 3).

Kaplan–Meier analysis was used to assess the patency, limb salvage and survival rates. In subgroup analyses, log rank test was used as the default method. However, this is appropriate when relative mortality/re-stenosis does not change with time. This is known as ‘‘proportional hazard’’. Immediately after SIA, the risk of complications is high initially violating the proportional hazard assumption. For this reason, we applied the Breslow test (Wilcoxon test) too, which gives more empha-sis to early deaths/events. Statistical analyempha-sis was performed with SPSS software (version 20.0; SPSS Inc, Chicago, IL, USA). The threshold of statistical signiﬁcance was set at 5%.

3. Results

42 infra-inguinal SIA procedures were performed in 37 patients during this period. Of these, 29 limbs on 26 patients fulﬁlled the inclusion criteria for the subset analysis of long occlusions. This comprised of 20 male and 6 female patients with a mean age of 69 years. 18 patients presented with inter-mittent claudication, 2 with rest pain and 9 had ulceration or gangrene (Table 4).

All treated lesions were infra-inguinal CTO, 23 exclusively SFA and 6 SFA-popliteal lesions. The mean length of the lesions was 27.3 cm (15.3–46.4). The mean SVS run-off scores were 3.6 (0–8) (Table 5).

Initial technical success was achieved in 27/29 of cases. The reason for both technically unsuccessful cases were due to fail-ure to break back into the true lumen. No re-entry devices were used in any of the cases. There were three major Class C complications in the form of distal thrombo-embolization which were successfully treated with suction thrombectomy and commenced on IV heparin. All 3 patients were admitted overnight, received IV heparin for 24–48 h and reported symp-tom improvement on follow-up. No minor complications were reported.

Table 3 Complications.

Deﬁnitions of complications Minor complications

A. No therapy, no consequence

B. Nominal therapy, no consequence, includes overnight admission for observation only

Major Complications

C. Require therapy, minor hospitalization (<48 h)

D. Require major therapy, unplanned increase in level of care, prolonged hospitalization (>48 h)

E. Permanent adverse sequelae F. Death

Table 4 Baseline patient demographics.

Age (years)

Median 69

Range 51–88

Gender (number of patients)

Male 20

Female 6

Risk factors (number of patients)

CKD3–5 (eGFR < 60) 5

Current smoker 15

Diabetes mellitus 7

Hypertension 20

Ischemic heart disease 18 Clinical category (fontaine classiﬁcation)

I (asymptomatic) 0

IIa (mild claudication) 0 IIb (moderate to severe claudication) 18 III (ischemic rest pain) 2 IV (ulceration or gangrene) 9 Concomitant medication (number of patients)

Statins 21

Anti-platelet therapy 18 Warfarin or phenindione 1

(5)

There were three deaths during the follow-up period, two from malignancies and one due to myocardial infarction more than 30 days post-procedure.

The mean duration of clinical follow-up was 414 days (median = 367 days).

Kaplan–Meier analysis showed a three-year cumulative proportional mortality rate of 82.5%.

With respect to limb salvage rate, there was only one below knee amputation in the study group giving a three-year cumu-lative proportional limb salvage rate of 96.6%. Following SIA, the target vessel thrombosed within 3 months. The patient went on to have a femoral-popliteal bypass graft which throm-bosed three times within a year (Table 6).

3.1. Primary patency

Case processing summary

Total N Nof events Censored

N Percent (%)

29 17 12 41.4

Life tablea

Interval start time

0 6 12 18 24

Number exposed to risk 27.5 18.0 6.5 2.0 1.0

a_{The median survival time is 10.445.}

3.2. Primary assisted patency

N Percent (%)

29 12 17 58.6

Life tablea

Interval start time

0 6 12 18 24

a_{The median survival time is 24.000.}

Table 5 Lesion characteristics.

Lower limb

Left 14

Right 15

Break back point (number of limbs)

Above joint line – SFA 23 Below joint line SFA/popliteal artery 6 Length of occlusion (centimeter)

Median 27.3

Range 15.3–46.4

SVS run-oﬀ score

Median 3.6

Range 0–8

Reason for stent deployment (number of patients) 8/29

Recoil 7

Fresh thrombus 1

ABPI

Median pre-procedure ABPI 0.59 Median post-procedure ABPI (6 weeks) 0.90 Median ABPI change 0.33

(6)

3.3. Secondary patency

N Percent (%)

29 11 18 62.1

Life tablea

Interval start time

0 6 12 18 24

a

The median survival time is 24.000.

N Percent (%)

29 3 26 89.7

Table 6 Outcomes.

Technical success 93%

Clinical success 93%

Duration of clinical follow-up (months)

Mean 13.8 Median 12.2 Range 2.8–34.5 Symptom recurrence Median (months) 12.6 Range 0–19.3 Primary patency Median 8.4 Range 0–34.5

Time to next intervention (months)

Median 8.4

Range 3.4–16.7

Time to occlusion (months)

Median 11.5

Range 0–29.9

Patient survival (months)

Median 31.3

Range 6.7–39.9

Limb salvage rate (months)

Median 31.3

Range 3–39.9

Complications (number of limbs)

Minor 0

Major 3

(7)

N Percent (%)

29 1 28 96.6

3.4. Comparisons

Overall comparisons

Chi-square df Sig. Log rank (Mantel-Cox) .637 1 .425 Breslow (generalized Wilcoxon) .380 1 .538 Test of equality of survival distributions for the different levels of critical ischemia.

Critical ischemia Total N Nof events Censored N Percent (%) No 18 8 10 55.6 Yes 11 3 8 72.7 Overall 29 11 18 62.1 Overall comparisons Chi-square df Sig. Log rank (Mantel-Cox) 1.614 2 .446 Tarone-Ware 1.425 2 .490

Test of equality of survival distributions for the different levels of no of calf vessels.

Case processing summary No of calf

vessels

N Percent (%)

1 4 2 2 50.0

2 8 4 4 50.0

3 17 5 12 70.6

(8)

Overall comparisons

Chi-square df Sig. Log rank (Mantel-Cox) 2.917 1 .088 Breslow (generalized Wilcoxon) 4.033 1 .045

Smokers Total N Nof events Censored

N Percent (%)

0 14 3 11 78.6

1 15 8 7 46.7

Overall 29 11 18 62.1

4. Discussion

During the study period, our institution performed 42 infra-inguinal SIA for CTO, of which 29 had disease in the TASC C and D category. Venous bypass has traditionally been con-sidered the ﬁrst line treatment for these lesions with reported 5 year patency rate of up to 69%. However, this is reliant on the availability of good quality veins and outﬂow popliteal or infra-geniculate arteries. When good veins are not available for bypass, it is debatable if prosthetic graft bypass is superior to SIA especially in CLI. As an institution, we have a low threshold to attempt SIA in patients with severe claudication or critical ischemia before considering bypass surgery in patients with multiple co-morbidities as there is an increased risk of perioperative mortality.

In 2007, Ko et al. compared SIA vs. intraluminal angio-plasty of long femoro-popliteal occlusions with an average length of 22.7 ± 9.9 cm. The primary patency of the SIA group on an intention-to-treat basis was 72.4% at 1 year. In their study, the entry site of all SIA cases was stented and patients were on a combination of aspirin and cilostazol or

dual antiplatelet therapy for at least one year. Patients with poor distal run-off were sometimes started on warfarin at the operator’s discretion. This practice is considerably different from our center where patients are only on dual antiplatelet therapy for 6 weeks and none were started on warfarin for poor run-offs. We have never prescribed cilostazol for our patients with peripheral vascular disease, but there is increas-ing evidence that cilostazol may reduce the risk of stenosis especially in patients with high risk of re-stenosis without increased risk of bleeding(2).

In 2010, Taneja et al. reported their results of 49 long length SFA and popliteal artery stenting in diabetic patients with a mean CTO of 20 cm. The primary patency rate at one year was low at 27% but the limb salvage rate at 1 year was accept-able at 80%. There were only 6 patients with diabetes in our cohort, thus only limited conclusions can be drawn from com-paring the two studies. However, there is a similar trend of low primary patency rates and multiple re-interventions are required in patients with TASC C and D disease, but amputa-tions are successfully prevented.

Dimitris et al. reported that CTO longer than 20 cm had an angiographic binary re-stenosis hazard ratio of 4.55 compared to lesions less than 20 cm. Other factors reported to negatively affect binary angiographic stenosis include initial CLI symp-toms and full lesion stenting (3). They found that selective entry/re-entry stenting was linked to lower re-stenosis rates. In the subgroup of initial CTO more than 20 cm, the re-stenosis rate at 1 year was more than 60%.

We have detected a significant difference in secondary patency between smoker and non-smokers (48% vs. 73% at 24 months, p = 0.045). It is expected that smokers will have poorer patency rates than non-smokers but we did not think that there will be such a significant difference between the two groups. The Basil Trial showed a hazard ratio of 1.72 for ex-smokers and 1.5 for current smokers in their multi-variate analysis (4). This hazard ratio was linked to survival rather than patency rates. There were more patients with crit-ical limb ischemia and diabetes within the non-smoking group suggesting that these confounding factors should reflect a poorer patency rate in this group and yet it fared better than the smoking group. Unfortunately, a more detailed smoking history was not obtained during initial patient assessment. We believe that this is required for future studies to determine which smoking associated factors affected patency rates, e.g. number of pack years, type of cigarettes, etc.

We recognize there are limitations in our single-center study with a small study group with heterogeneity of critical and non-critical limbs. Although our sample size is small, the evidence in the literature on TASC C and D CTO is currently limited. We believe that this study may be a valuable addition to the body of evidence that demonstrates SIA is a safe and effective alternative for long severe CTO. This remains with the caveat that a rigorous surveillance program and a re-intervention policy are in place to ensure reasonable rates of primary assisted patency.

5. Conclusion

In conclusion, SIA is effective alternative in treating TASC C and D CTOs with limb salvage and survival rates comparable to surgery. Although the re-intervention rates are high, it

(9)

remains a viable treatment option especially in high risk surgi-cal candidates with multiple co-morbidities. Smoking decreases patency rates and more detailed smoking history is required in future studies to determine which smoking associ-ated factor is responsible.

6. Conﬂict of interest

None declared.

References

(1)Peterkin GA, Manabe S, LaMorte WW, Menzoian JO. Evaluation of a proposed standard reporting system for preoperative angiograms in infrainguinal bypass procedures: angiographic

correlates of measured runoff resistance. J Vasc Surg

1988;7(3):379–85 [Epub 1988/03/01].

(2)Ding XL, Xie C, Jiang B, Gao J, Zhang LL, Zhang H, et al. Efﬁcacy and safety of adjunctive cilostazol to dual antiplatelet therapy after stent implantation: an updated meta-analysis of randomized controlled trials. J Cardiovasc Pharmacol Ther 2013;18(3):222–8 [Epub 2012/12/25].

(3)Siablis D, Diamantopoulos A, Katsanos K, Spiliopoulos S, Kagadis GC, Papadoulas S, et al. Subintimal angioplasty of long chronic total femoropopliteal occlusions: long-term outcomes, predictors of angiographic restenosis, and role of stenting. Cardiovasc Interventional Radiol 2012;35(3):483–90 [Epub 2011/ 08/13].

(4)Bradbury AW, Adam DJ, Bell J, Forbes JF, Fowkes FG, Gillespie I, et al. Bypass versus angioplasty in severe ischaemia of the leg (BASIL) trial: a survival prediction model to facilitate clinical decision making. J Vasc Surg 2010;51(5 Suppl):52S–68S [Epub 2010/05/15].